Frictional Strength Heterogeneity

1
Frictional Strength Heterogeneity Surface Heat
flow near SAFOD(Or, Is the creeping section
strong?)

• Matthew A. dAlessio
• Colin F. Williams
• Roland Bürgmann, UC Berkeley

U.S. Department of the Interior U.S. Geological
Survey
2
IF FAULTS ARE WEAK
What mechanism weakens faults?

• Dynamic Weakening (related to earthquake rupture
  process)
• Thermal pressurization of fluids
• Chattering during slip
• Acoustic fluidization of gouge

• Permanent Weakness
• Weak gouge material
• Serpentine?
• Some Clays?
• Constant fluid pressure from (deep) fluid source

http//www.nps.gov/prsf/geologyy/serpent.htm
3
TEST CASE CREEPING SECTION

• 170-km-long section of San Andreas in central
  California
• Slip almost entirely by slow, aseismic creep
• No Large Earthquakes
• No Dynamic Weakening

Creeping Section
(Base Figure SAFOD Team)
4
CREEPING SECTION HEAT FLOW DATA GAP
SAFOD
5
IS THE CREEPING SECTION STRONG?

• Thanks to Jim Brune for raising this question.
• Creeping section IS a frictional asperity.
• Aseismic creep controlled by velocity
  strengthening frictional behavior.
• Velocity strengthening ? strong
• But high ? minerals that promote creep do exist
• Heat flow data are sparse in creeping section.

6
FAULT STRENGTH HETEROGENEITY

• PROBLEM!
• All existing models of frictional heat assume
  infinitely long, homogeneous faults
• Faults are not frictionally uniform
• Asperities
• Special frictional behavior along strike
• Creeping section of SAF

7

• Single strong asperity (stationary)
• Midpoint heat flow similar to infinite case
• Heat flow at tip is half that of midpoint

8
MOVING ASPERITY
As fault slips, block offset transports heat (and
possibly asperity) along strike.
9

• Moving Asperity in Action.

10

• Strong moving asperity
• Heat flow asymmetric
• Heat flow tail at trailing edge

Infinite fault
This Case
11
SLIP RATE DEPENDENCE
Faster slip rate more asymmetry
12
OTHER SCENARIOS
13
OTHER SCENARIO RESULTS
See Poster Manuscript
14
OTHER SCENARIO RESULTS
Heat Flow mW/m2
High friction asperity Dragging scenario

• Heat flow NOT correlated with distance from
  fault.
• (Lachenbruch Sass, 1980)

15
IS THE CREEPING SECTION STRONG?

• Probably not.
• Sense of asymmetry in heat flow data opposite
  that of model predictions for moving asperity.
• Heat flow not correlated with distance from
  fault.
• Weak Creeping Section CANNOT be weakened
  dynamically.
• It doesnt slip in large earthquakes!
• Permanent weakness must weaken the creeping
  section.
• Dynamic weakness alone cannot explain weak faults.

Dynamic Weakening
16
The End
U.S. Department of the Interior U.S. Geological
Survey
17
CONCLUSIONS

• Are faults frictionally weak?
• Probably, according to limited heat flow and
  thermochronology.
• Have we ever observed frictional heat?
• Not yet. Always below our detection threshold.
• Weakening Mechanisms?
• San Gabriel fault Dont necessarily require
  abundant fluids.
• Creeping section Dont necessarily require
  dynamic processses.
• MULTIPLE MECHANISMS (?)

18
SAFOD THERMOCHRONOLOGY

• Use (U-Th)/He Fission-track thermochronology
• Reconstruct long-term thermal history near San
  Andreas fault

19
THERMAL HISTORY

• Heating Event
• Starts 30 Ma
• Onset of SAF
• Lasts until 5 Ma
• Change in SAF
• Burial, Volcanic, or FRICTIONAL HEAT?

20
TARGET SAFOD

• Known depth
• Known thermal history
• Surface heat flow
• Thermochronology
• Major plate boundary
• East side of fault traveled from locked section.
• Fault zone width?
• Sample availability?

(Figure SAFOD Team)